A Reliable and Repeatable Model for Predicting Microvascular Invasion in Patients With Hepatocellular Carcinoma.

RATIONALE AND OBJECTIVES: The reproducibility of imaging models for predicting microvascular invasion (MVI) in patients with hepatocellular carcinoma (HCC) remains questionable due to inconsistent interpretation of image signs. Our aim was to screen for high-consensus MRI features to develop a repeatable model for predicting MVI. MATERIALS AND METHODS: We included 219 patients with HCC who underwent surgical resection, and patients were divided into a training cohort (n = 145) and a validation cohort (n = 74). Morphological characteristics, signal features on hepatobiliary phases, and dynamic enhancement patterns were qualitatively interobserver evaluated. Interobserver agreement was assessed using Cohen's κ for selecting features with high interobserver agreement. Risk factors that were significant in stepwise multivariate analysis and that could be measured with good interobserver agreement were used to construct a predictive model, which was assessed in the validation cohort. The diagnostic performance of the model was evaluated based on area under the receiver operating characteristic curve (AUC). RESULTS: Multivariate analysis identified nonsmooth tumor margin, absence of radiologic capsule, and intratumoral artery as independent risk factors of MVI. These MRI-based features showed good or nearly perfect interobserver agreement between radiologists (κ > 0.6). The predictive model predicted MVI well in the training (AUC 0.734) and validation cohorts (AUC 0.759) and fitted well to calibration curves. CONCLUSION: MRI features included nonsmooth tumor margin, absence of radiologic capsule, and intratumoral artery that can be assessed with high interobserver agreement can predict MVI in HCC patients. The predictive model described here may be useful to radiologists, regardless of experience level.

Diagnostic value of diffusion-weighted magnetic resonance imaging for local and skull base recurrence of nasopharyngeal carcinoma after radiotherapy.

Tumor recurrence is a major cause of nasopharyngeal carcinoma (NPC) treatment failure. Diffusion-weighted imaging (DWI) is used for a variety of cancers, but few data are available for NPC.The aim of the study was to investigate the DWI features of recurrent NPC after radiotherapy and apparent diffusion coefficient (ADC) thresholds for the diagnosis of recurrent NPC.This was a retrospective study of 160 patients with NPC treated by radiotherapy at the Cancer Hospital affiliated to Guangxi Medical University from May 2012 to March 2015. The patients were divided into the local recurrence (n = 39), fibrosis (n = 51), clivus recurrence (n = 22), and clivus nonrecurrence (n = 48) groups. The patients underwent magnetic resonance imaging (MRI), enhanced MRI, and DWI. Receiver operating characteristics curves were used to determine sensitivity, specificity, and negative predictive values.ADC values were significantly different between the recurrence and fibrosis groups (P < .0001). Using ADC threshold values of 0.887 × 10 mm/s for local recurrence, the area under the curve (AUC) of DWI was 0.967 (87.2% sensitivity and 94.1% specificity), compared with 0.732 for routine MRI (71.8% sensitivity and 74.5% specificity) (P < .001). Using ADC threshold values of 1.018 × 10 mm/s for the diagnosis of clivus recurrent NPC, the AUC of DWI was 0.984 (95.5% sensitivity and 91.7% specificity) compared with 0.558 for routine MRI (63.6% sensitivity and 47.9% specificity) (P < .001).DWI has a higher diagnostic value for recurrent NPC than MRI. DWI can increase the diagnosis sensitivity and specificity of locally recurrent NPC.

Diffusion-weighted MR imaging in prediction of response to neoadjuvant chemotherapy in patients with breast cancer.

This study aims to evaluate the potential of apparent diffusion coefficient (ADC) derived from diffusion-weighted MR imaging for predicting the treatment response to neoadjuvant chemotherapy (NACT) in patients with breast cancer. Magnetic resonance imaging was performed prior to NACT and after two cycles of NACT. The correlation between mean ADCpre values, mean ADCpost values, changes in ADC values and changes in tumor diameters after NACT was examined using Spearman rank correlation. A total of 164 breast cancers were enrolled in this study. Mean ADCpre values of responders ([0.85 ± 0.16] × 10-3 mm2/s) and non-responders ([0.84 ± 0.21] × 10-3 mm2/s) had no significant difference (P = 0.759). While mean ADCpost value of responders was significantly higher than that of non-responders ([1.17 ± 0.37] × 10-3 mm2/s vs. [1.01 ± 0.28] × 10-3 mm2/s; P = 0.002). Both mean ADCpost values (r = 0.288, P = 0.000) and changes in mean ADC values (r = 0.222, P = 0.004) were positively correlated to changes in tumor diameter after NACT, except for mean ADCpre values (r = 0.031, P = 0.695). Our results indicated that mean ADCpost values and changes in ADC values after NACT might be a biological marker for assessing the efficacy of chemotherapy.

Validated preoperative computed tomography risk estimation for postoperative hepatocellular carcinoma recurrence.

AIM: To develop and validate a risk estimation of tumor recurrence following curative resection of operable hepatocellular carcinoma (HCC). METHODS: Data for 128 patients with operable HCC (according to Barcelona Clinic Liver Cancer imaging criteria) who underwent preoperative computed tomography (CT) evaluation at our hospital from May 1, 2013 through May 30, 2014 were included in this study. Follow-up data were obtained from hospital medical records. Follow-up data through May 30, 2016 were used to retrospectively analyze preoperative multiphasic CT findings, surgical histopathology results, and serum α-fetoprotein and thymidine kinase-1 levels. The χ2 test, independent t-test, and Mann-Whitney U test were used to analyze data. A P-value of < 0.05 was considered statistically significant. RESULTS: During the follow-up period, 38 of 128 patients (29.7%) had a postoperative HCC recurrence. Microvascular invasion (MVI) was associated with HCC recurrence (χ2 = 13.253, P < 0.001). Despite postoperative antiviral therapy and chemotherapy, 22 of 44 patients with MVI experienced recurrence after surgical resection. The presence of MVI was 57.9% sensitive, 75.6% specific and 70.3% accurate in predicting postoperative recurrence. Of 84 tumors without MVI, univariate analysis confirmed that tumor margins, tumor margin grade, and tumor capsule detection on multiphasic CT were associated with HCC recurrence (P < 0.05). Univariate analyses showed no difference between groups with respect to hepatic capsular invasion, Ki-67 proliferation marker value, Edmondson-Steiner grade, largest tumor diameter, necrosis, arterial phase enhanced ratio, portovenous phase enhanced ratio, peritumoral enhancement, or serum α-fetoprotein level. CONCLUSION: Non-smooth tumor margins, incomplete tumor capsules and missing tumor capsules correlated with postoperative HCC recurrence. HCC recurrence following curative resection may be predicted using CT.

GSTT1 and GSTM1 polymorphisms predict treatment outcome for breast cancer: a systematic review and meta-analysis.

Observational studies have reported controversial results on the association between GSTT1 and GSTM1 genotypes and treatment outcome of breast cancer. The purpose of this study is to evaluate the association between GSTT1 and GSTM1 and treatment outcome in breast cancer patients. Eligible studies were searched in PubMed, EMBASE, Cochrane Library, and China National Knowledge Infrastructure databases. A random-effect model or fixed-effect model was used to calculate the overall combined risk estimates. Twenty-one studies with a total of 4990 patients were included in this meta-analysis. The GSTM1 null genotype (odds ratio (OR) = 1.33, 95 % confidence interval (CI) 1.01-1.75, P = 0.046) and GSTT1/GSTM1 double null genotype (OR = 2.22, 95 % CI 1.02-4.84, P = 0.045) were significantly associated with an increased tumor response. A reduced overall survival (hazard ratio (HR) = 0.84, 95 % CI 0.72-0.98, P = 0.024) was observed in GSTM1 null genotype, especially in mixed descent (HR = 0.77, 95 % CI 0.61-0.96, P = 0.018) and large sample size (HR = 0.85, 95 % CI 0.72-0.99, P = 0.033). Evidence of publication bias was observed in GSTM1 genotype rather than in GSTT1 genotype. This meta-analysis suggests that GSTM1 null and GSTT1/GSTM1 double null polymorphisms might be significantly associated with an increased tumor response. However, the GSTM1 null genotype might be significantly associated with a reduced overall survival. Future studies are warranted to confirm these findings.

Diagnostic performance of ADCs in different ROIs for breast lesions.

OBJECTIVE: The purpose of this study was to explore the diagnostic performance of apparent diffusion coefficient (ADC) values for breast lesions by different measuring methods and find out the optimum measuring method. METHODS: ADCW-mean and ADCW-min were obtained by whole-measurement method, while ADCmean and ADCmin were extracted by spot-measurement method. Four ADCs were analyzed by One-way ANOVA and Independent T-test. The diagnostic performances of these four ADCs were calculated by receiver operating characteristics (ROC) curves and the area under the curves (AUC) were compared through Z-test. RESULTS: For the whole-measurement method, there were significant differences between malignant and benign lesions (ADCW-mean=1.014±0.197 for malignant, ADCW-mean=1.650±0.348 for benign, F=37.511, P<0.001; ADCW-min=0.627±0.144 for malignant, ADCW-min=1.245±0.290 for benign, F=41.446, P<0.001), as well as the spot-measurement method (ADCmean=1.010±0.234 for malignant, ADCmean=1.648±0.392 for benign, F=34.580, P<0.001; ADCmin=0.817±0.203 for malignant, ADCmin=1.411±0.357 for benign, F=40.039, P<0.001). The optimal diagnostic threshold of ADCW-mean, ADCW-min, ADCmean, and ADCmin values were 1.223×10(-3) mm(2)/s, 0.897×10(-3) mm(2)/s, 1.315×10(-3) mm(2)/s, and 1.111×10(-3) mm(2)/s, respectively. ROC curves indicated that the AUC for ADCW-min (0.969) was statistically significant higher than the AUC for ADCW-mean (0.940; Z=2.473, p=0.013), ADCmean (0.919; Z=3.691, P=0.000), and ADCmin (0.928; Z=3.634, P=0.000). The AUC for ADCW-mean was also significantly higher than the AUC for ADCmean (Z=2.863, P=0.004). CONCLUSION: The results provided evidence that the most reliable and accurate value in demonstrating the limitation of diffusion may be ADCW-min, and it has the highest diagnostic value in distinguishing breast lesions from malignant to benign.